This proposal is a continuation of project 2A in the previous SPORE cycle. Most invasive pancreatic adenocarcinomas are incurable. Detection of pancreatic neoplasia at the pre-invasive stages offers the best chance of cure. The objective of this proposal is to elucidate the molecular abnormalities in the three recognized precursor lesions of pancreatic adenocarcinoma - Pancreatic Intraepithelial Neoplasia (PanIN), Intraductal Papillary Mucinous Neoplasm (IPMN), and Mucinous Cystic Neoplasm (MCN). IPMNs and MCNs, in particular, are the most common underlying etiology for asymptomatic pancreatic cysts, and in concert with the increasing usage of sophisticated radiological scans, it is anticipated that thousands of such cysts will be identified each year. Clinicians do not yet have the means to accurately differentiate the neoplastic cysts (IPMN and MCN) from a plethora of non-neoplastic cysts, leading to unnecessary therapeutic intervention in many instances, and a missed opportunity for cure with others. By identifying molecular alterations in the precursor lesions of pancreatic adenocarcinoma, markers will be developed for translational application in clinical settings. The long term goal of this proposal is to alleviate pancreatic cancer mortality by facilitating the identification of pancreatic neoplasia at its earliest stages. Aim 1: To define genetic alterations in the precursors of invasive pancreatic adenocarcinoma (IPMNs, MCNs and PanlNs) in patients with sporadic and familial pancreatic cancers. Aim 2: To develop tissue-based markers for diagnosis and risk assignation in asymptomatic pancreatic cysts, and apply these markers to relevant clinical specimens, including pancreatic cyst fluid and fine needle aspiration cytology material. Aim 3: To translate the new markers from Aims 1 and 2 in expanded clinical settings, including rare variant tumors, for developing therapeutic strategies, and for non-invasive imaging. This project will utilize tissue materials accrued by SPORE Core 2 and collaborators, and will strongly interface with proposed Projects 2A, 3A, and 3B in order to translate marker discoveries to patient care. Lay summary: Pancreatic cancer, once invasive, is essentially incurable. The thrust of this project is to develop molecular strategies that can identify pancreatic cancer at its earliest stages, and therefore cure the patient. Non-invasive lesions that precede the development of pancreatic cancer will be studied, including small (microscopic) and radiologically detectable (macroscopic) lesions, with the intent of developing clinically useful tests that can identify pancreatic cancer as early as possible.